Using pandas.IndexSlice to drop rows with MultiIndex

Question

Recently I wanted to use a pd.IndexSlice to drop some rows but realised they don't work together.

Here's an example:

data = [('animal', 'cat', 'black', 1),
         ('animal', 'mouse', 'grey', 5),
         ('plant', 'houseplant', 'green', 11)]

df = pd.DataFrame(data, columns=['type','species', 'colour', 'amount'])

df = df.set_index(['type','species','colour'])

idx = pd.IndexSlice

ind = idx[:,'mouse']

From this code we get a multi-indexed dataframe.

df

                              amount
type   species    colour        
animal cat        black        1
       mouse      grey         5
plant  houseplant green       11

if you access that row index it will output:

df.loc[ind,:]

                       amount
type   species colour        
animal mouse   grey         5

But if I want to use the same IndexSlice to drop some rows you will find out that it doesn't allow slicing:

df.drop(ind)

For this you'll get an error: TypeError: unhashable type: 'slice'

How can I use the IndexSlice to drop that row?

Federico Garza · Accepted Answer · 2018-09-10 11:48:33Z

1

This is my current solution to this problem. Please comment and add your solutions.

Dropping works with pd.IndexSlice by running the slicer through the dataframe and extracting the index, then using that to drop the row:

df.drop(df.loc[ind,:].index)

                          amount
type   species    colour        
animal cat        black        1
plant  houseplant green       11

answered Sep 10, 2018 at 11:48

Federico Garza

1571 gold badge1 silver badge11 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Using pandas.IndexSlice to drop rows with MultiIndex

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related